Milos Bozic

Learn More
Providing accurate load forecast to electric utility corporations is essential in order to reduce their operational costs and increase profits. Hence, training set selection is an important preprocessing step which has to be considered in practice in order to increase the accuracy of load forecasts. The usage of mutual information (MI) has been recently(More)
This paper presents a model for short-term load forecasting using least square support vector machines. Available data are analyzed and appropriate features are selected for the model. Last 24 hours load demands are used for features in combination with day in week and hour in day. It is shown that temperature is not always a very good feature for the(More)
In the deregulated energy market, the accuracy of load forecasting has a significant effect on the planning and operational decision making of utility companies. Electric load is a random non-stationary process influenced by a number of factors which make it difficult to model. To achieve better forecasting accuracy, a wide variety of models have been(More)
This paper presents the maximally balanced connected partition (MBCP) problem in graphs. MBCP is to partition a weighted connected graph into the two connected subgraphs with minimal misbalance, i.e., the sums of vertex weights in two subgraphs are as much equal as possible. The MBCP has many applications both in science and practice, including education.(More)
This paper presents an approach for the medium-term load forecasting using Support Vector Machines (SVMs). The proposed SVM model was employed to predict the maximum daily load demand for the period of a month. Analyses of available data were performed and the most important features for the construction of SVM model are selected. It was shown that the size(More)
Training set instance selection is an important preprocessing step in many machine learning problems, including time series prediction, and has to be considered in practice in order to increase the quality of the predictions and possibly reduce training time. Recently, the usage of mutual information (MI) has been proposed in regression tasks, mostly for(More)
Logistic regression (LR) approach for power transformer fault diagnosis, based on dissolved gas analysis (DGA) is presented in this paper. DGA methods proposed by actual standard IEC 60599, often identify wrong fault or cannot even recognize fault type. To overcome these problems, in recent years, several artificial intelligence (AI) approaches are(More)
  • 1